A public service data analysis system based on artificial intelligence

Abstract

The present invention discloses a public service data analysis system based on artificial intelligence, which relates to the field of data analysis technology, and includes a data processing module, which extracts data features based on public service data and personal user information; and a data analysis module, which constructs a regression model and analyzes public service data to predict environmental scores. The present invention provides a basis for predicting environmental scores of public places through the collection and standardized processing of public service data. By combining environmental scores, user behavior intensity, and the interaction effect between the two, the personalized behavior patterns and environmental impacts of users are comprehensively considered, and the consideration of personal travel intentions is retained. At the same time, based on public service data, the travel impacts caused by traffic and environmental factors can be reflected in the comprehensive score in a timely manner, achieving the effect of comprehensively analyzing public service data and identifying user behaviors based on artificial intelligence, and providing users with travel predictions and recommendations.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a public service data analysis system based on artificial intelligence. Background Art

[0002] Public service data, including environmental data, traffic data, social activity data, etc., can provide strong decision-making support for public service agencies through in-depth analysis of these data. In cities, people usually choose to travel according to their personal schedules. In addition to people's travel intentions, the quality of their travel experience is also easily affected by environmental data and traffic data.

[0003] Although artificial intelligence technology has broad application prospects in public service data analysis, it has not yet been combined with the city's public service data, especially in the use of environmental data and traffic data, which makes it inconvenient to conduct comprehensive analysis and provide effective suggestions for pedestrians' travel. As a result, it is impossible to provide people with accurate and timely predictions and recommendations in times of traffic congestion and poor environment, which affects pedestrian travel and increases the pressure on traffic and environmental public services. Summary of the invention

[0004] In view of the above-mentioned problems existing in the existing public service data analysis system based on artificial intelligence, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is that it is inconvenient to conduct comprehensive analysis in the use of environmental data and traffic data, and to provide effective suggestions for pedestrians' travel. As a result, it is impossible to provide people with accurate and timely predictions and recommendations in time when traffic is congested and the environment is bad, which affects pedestrian travel and increases the pressure on traffic and environmental public services.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a public service data analysis system based on artificial intelligence, comprising:

[0007] Data processing module, based on public service data and personal user information, and extracting data features;

[0008] Data analysis module, builds regression models, analyzes public service data to predict environmental scores, analyzes user travel behavior based on personal user information, and analyzes user behavior intensity;

[0009] Comprehensive scoring module, which integrates environmental scores and user behavior intensity, analyzes the comprehensive score of travel, dynamically optimizes thresholds, and makes travel recommendations based on the comprehensive score;

[0010] The data management module performs tiered storage and backup based on the cloud service storage platform, automatically generates reports based on stored data, and updates data regularly.

[0011] As a preferred solution of the public service data analysis system based on artificial intelligence of the present invention, wherein: the public service data and personal user information are used to extract feature data, including:

[0012] Based on the feature extraction of environmental data and traffic data in public service data, including PM2.5, temperature, humidity, ultraviolet intensity and traffic flow data, the PM2.5 is logarithmically changed, and the temperature, humidity, ultraviolet intensity and traffic flow data are standardized to form the environmental feature vector E;

[0013] Collect personal user information, including user location information at different times, travel displacement data, and historical travel time data, and perform standardized processing to form a user feature vector R.

[0014] As a preferred solution of the public service data analysis system based on artificial intelligence described in the present invention, wherein: the construction of a regression model to analyze public service data and predict environmental scores includes:

[0015] Based on the environmental feature vector E, a regression model is constructed to perform environmental scoring for the target public service scenario, which is expressed as:

[0016] ;

[0017] in represents the predicted environmental score, Indicates the upper limit of the frequency division time interval, represents the weight coefficient of the i-th environmental feature, represents the adjustment coefficient of the i-th environmental feature, represents the frequency coefficient of the i-th feature, represents the i-th eigenvalue at time t, and N represents the number of features;

[0018] Use the training set to train the model and set the mean square error loss function;

[0019] Use gradient descent to minimize the loss function to update and determine the parameters , and ;

[0020] Use the newly collected feature data to input the model to obtain the current environmental score.

[0021] As a preferred solution of the public service data analysis system based on artificial intelligence of the present invention, wherein: the analysis of user travel behavior based on personal user information and the analysis of user behavior intensity include:

[0022] Based on the user feature vector R, the user behavior intensity is calculated and expressed as:

[0023] ;

[0024] in represents the behavior intensity of the i-th user at time t, represents the adjustment coefficient of the intensity of the periodic behavior of the i-th user, T represents the time period, L represents the reference distance of the permanent location, Indicates the distance between the current location of user i and the permanent location, represents the time mean of user i’s current behavior, represents the time standard deviation of user i’s current behavior, represents the adjustment coefficient of user i’s historical action data, represents the time mean of the k-th historical behavior, represents the time standard deviation of the k-th historical behavior, is the length of the time period, K represents the total number of historical behavior events;

[0025] By collecting historical data and calibrating the historical user behavior intensity data based on user behavior as training set data, setting the mean square error loss function, using the gradient descent method to minimize the loss function, adjusting and confirming and ;

[0026] Use the newly collected user feature data to predict the intensity of user behavior.

[0027] As a preferred solution of the public service data analysis system based on artificial intelligence of the present invention, wherein: the comprehensive environment score and user behavior intensity are used to analyze the comprehensive score of the trip, including:

[0028] According to the predicted environment score and user behavior intensity, the comprehensive score is calculated and expressed as:

[0029] ;

[0030] Where S represents the comprehensive score, represents the environmental score adjustment coefficient, represents the user behavior intensity adjustment coefficient, The interaction coefficient between the environment score and the user behavior intensity;

[0031] By collecting historical data, we define historical comprehensive scores as training sets based on different weather environment data and users' actual behaviors, use mean square error as the objective function and define the parameter range, and divide the data set into k subsets through k-fold cross validation. Each time, we use k-1 subsets for training and the remaining subset for validation. This is repeated k times, and the average value is taken as the value of the objective function. The value that minimizes the objective function is used to confirm the result. , and ;

[0032] Using the newly predicted environment score and the intensity of the user's behavior, a comprehensive score is calculated.

[0033] As a preferred solution of the public service data analysis system based on artificial intelligence of the present invention, wherein: the dynamic optimization threshold, judging travel recommendation according to the comprehensive score, includes:

[0034] Calculate the statistics of the comprehensive score S in the user's historical travel data, including the mean, standard deviation and median, and add the mean and standard deviation as the preliminary threshold ;

[0035] Based on historical travel data, the travel ratio TPR of actual travel users that are correctly predicted is calculated, and the non-travel ratio TNR of non-travel users that are correctly predicted is calculated;

[0036] Adjust the threshold gradually And verify that the correct prediction ratio of TNR and TPR reaches the maximum, determine the threshold ;

[0037] According to the determined threshold If the comprehensive score S is greater than the threshold, the user is recommended to travel. If the comprehensive score S is less than or equal to the threshold, the user is not recommended to travel.

[0038] As a preferred solution of the public service data analysis system based on artificial intelligence of the present invention, wherein: the hierarchical storage and backup based on the cloud service storage platform includes:

[0039] The historical data, user behavior data and comprehensive score data are remotely transmitted to the cloud service storage platform via wireless network, and the data is regularly backed up through the AWS S3 formula;

[0040] Use cloud storage services to store data separately and adopt a tiered storage strategy based on data access frequency and importance;

[0041] AWS S3 is used as cold data storage for historical data and backup data, and AWS S3 is used as hot data storage for real-time comprehensive scores and user behavior data.

[0042] As a preferred solution of the public service data analysis system based on artificial intelligence of the present invention, wherein: the automatic generation of reports based on stored data and regular data updates include:

[0043] Use AWS QuickSight based on the cloud service platform as a report generation tool, integrate the data in the cloud storage with the report generation tool, and build data models and reports;

[0044] Use AWS QuickSight to configure the automatic refresh policy of the SPICE dataset to synchronize data regularly and configure the automatic update function of the report generation tool to refresh data and reports regularly.

[0045] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of the above-mentioned artificial intelligence-based public service data analysis system when executing the computer program.

[0046] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned artificial intelligence-based public service data analysis system.

[0047] The beneficial effects of the present invention are as follows: by collecting and standardizing public service data, a basis is provided for predicting environmental scores of public places; by combining environmental scores, user behavior intensity and the interaction effect between the two, the user's personalized behavior patterns and environmental impacts are comprehensively considered, and personal travel intentions are retained; at the same time, based on public service data, the travel impacts caused by traffic and environmental factors can be reflected in the comprehensive score in a timely manner, achieving the effect of providing users with travel predictions and recommendations based on comprehensive analysis of public service data and identification of user behavior based on artificial intelligence, providing users with psychological expectations for travel, reducing the impact on user pedestrians, and reducing the pressure on traffic and environmental public services. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0049] Figure 1 This is a schematic diagram of the structure of the public service data analysis system based on artificial intelligence.

[0050] Figure 2 This is a flowchart of the public service data analysis system based on artificial intelligence. DETAILED DESCRIPTION

[0051] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0052] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0053] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0054] Reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and which provides a public service data analysis system based on artificial intelligence. The public service data analysis system based on artificial intelligence includes:

[0055] Data processing module, based on public service data and personal user information, and extracting data features;

[0056] Preferably, based on public service data and personal user information, and extracting feature data, including,

[0057] Based on the feature extraction of environmental data and traffic data in public service data, including PM2.5, temperature, humidity, ultraviolet intensity and traffic flow data, the PM2.5 is logarithmically changed, and the temperature, humidity, ultraviolet intensity and traffic flow data are standardized to form the environmental feature vector E;

[0058] Collect personal user information, including user location information at different times, travel displacement data, and historical travel time data, and perform standardized processing to form a user feature vector R.

[0059] Through the collection and standardization of public service data, a basis is provided for the prediction of environmental scores for public places. Through the collection and standardization of data based on personal user information, a basis is provided for the prediction of user behavior intensity. The data standardization process makes data of different dimensions have a consistent scale, which is conducive to the comparison and comprehensive analysis of different features, eliminates dimensional differences in the data, and reduces analytical errors caused by different data magnitudes. At the same time, the logarithmic change processing of PM2.5 can reduce data volatility and improve data stability and analysis accuracy.

[0060] Data analysis module, builds regression models, analyzes public service data to predict environmental scores, analyzes user travel behavior based on personal user information, and analyzes user behavior intensity;

[0061] Preferably, a regression model is constructed to analyze public service data and predict environmental scores, including:

[0062] Based on the environmental feature vector E, a regression model is constructed to perform environmental scoring for the target public service scenario, which is expressed as:

[0063] ;

[0064] in represents the predicted environmental score, Indicates the upper limit of the frequency division time interval, represents the weight coefficient of the i-th environmental feature, represents the adjustment coefficient of the i-th environmental feature, represents the frequency coefficient of the i-th feature, represents the i-th eigenvalue at time t, and N represents the number of features;

[0065] Use the training set to train the model and set the mean square error loss function, expressed as:

[0066] ;

[0067] Where M represents the number of samples, represents the calculation loss, represents the actual environmental score of the jth sample, represents the value of the i-th feature of the j-th sample at time t;

[0068] Use gradient descent to minimize the loss function to update and determine the parameters , and ;

[0069] Use the newly collected feature data to input the model to obtain the current environmental score;

[0070] The model formula is based on the basic regression model formula: , introduced logarithmic transformation, sine function and integration operation. The PM2.5 data was logarithmically transformed to make the data distribution more normal, reduce the impact of outliers, and improve the robustness of the model. The temperature, ultraviolet intensity and humidity were standardized so that the eigenvalues ​​were in the same dimension, which was convenient for model processing and calculation. The integration operation ensured that the scoring formula could reflect the changing trend of the data in the entire time interval. The environmental characteristics at different time points were comprehensively considered. The sine function was introduced to reflect the periodic changes of the data, especially the daily and seasonal changes in the environmental data.

[0071] By considering environmental data from a variety of public service data, we comprehensively evaluate the impact of environmental conditions on travel to the destination, especially the historical scores of the destination, and the mutual influence between the environmental conditions and weather factors. By weighting and transforming the characteristics of each factor, we ensure that each factor makes a reasonable contribution to the score. This comprehensive processing can provide a more comprehensive and accurate environmental score, helping users make more informed decisions. By introducing sinusoidal functions and integral operations, we can effectively reflect the dynamic changes of environmental data, so that the scoring system can more accurately reflect the real-time changes of environmental conditions and provide users with more reliable scoring results. Based on the environmental scoring results, the system can provide users with more reasonable and scientific travel recommendations.

[0072] Furthermore, based on personal user information, the user's travel behavior is analyzed, and the intensity of the user's behavior is analyzed, including:

[0073] Based on the user feature vector R, the user behavior intensity is calculated and expressed as:

[0074] ;

[0075] in represents the behavior intensity of the i-th user at time t, represents the adjustment coefficient of the intensity of the periodic behavior of the i-th user, T represents the time period, L represents the reference distance of the permanent location, Indicates the distance between the current location of user i and the permanent location, represents the time mean of user i’s current behavior, represents the time standard deviation of user i’s current behavior, represents the adjustment coefficient of user i’s historical action data, represents the time mean of the k-th historical behavior, represents the time standard deviation of the k-th historical behavior, is the length of the time period, K represents the total number of historical behavior events;

[0076] By collecting historical data and calibrating the historical user behavior intensity data based on user behavior as the training set data, the mean square error loss function is set, which is expressed as:

[0077] ;

[0078] in Indicates the behavior intensity of user behavior based on historical data. Indicates the predicted behavioral strength;

[0079] Use gradient descent to minimize the loss function, adjust and confirm and ;

[0080] Use the newly collected user feature data to predict the user's behavior intensity;

[0081] Formula using the sine function To simulate the user's daily behavior cycle, and adjust the behavior intensity in combination with the geographical displacement, to reflect the user's behavior changes in different geographical locations, use the Gaussian distribution function It indicates the concentration of user behavior in a specific time period and reflects the time distribution characteristics of the behavior. It is calculated through the weighted sum of historical behavior data. Reflects the impact of historical behavior on current behavior. The impact of each historical behavior event is weighted according to the time and intensity of its occurrence.

[0082] Through the setting of the model, the user's periodic behavior, changes in geographical location and historical behavior data are comprehensively considered, so as to more accurately predict the user's behavior intensity at different times and geographical locations. By adjusting the geographical displacement and adjustment coefficient, the model can reflect the behavioral habits of each user in a personalized way, making the recommendation system and behavior prediction more accurate. The model can be dynamically adjusted as new behavioral data and geographical displacement change, improving the real-time and accuracy of the prediction.

[0083] Comprehensive scoring module, which integrates environmental scores and user behavior intensity, analyzes the comprehensive score of travel, dynamically optimizes thresholds, and makes travel recommendations based on the comprehensive score;

[0084] Preferably, the comprehensive score of the trip is analyzed by combining the environmental score and the user behavior intensity, including:

[0085] According to the predicted environment score and user behavior intensity, the comprehensive score is calculated and expressed as:

[0086] ;

[0087] Where S represents the comprehensive score, represents the environmental score adjustment coefficient, represents the user behavior intensity adjustment coefficient, The interaction coefficient between the environment score and the user behavior intensity;

[0088] By collecting historical data, we define historical comprehensive scores as training sets based on different weather environment data and users' actual behaviors, use mean square error as the objective function and define the parameter range, and divide the data set into k subsets through k-fold cross validation. Each time, we use k-1 subsets for training and the remaining subset for validation. This is repeated k times, and the average value is taken as the value of the objective function. The value that minimizes the objective function is used to confirm the result. , and ;

[0089] Using the newly predicted environment score and the user behavior intensity, calculate a comprehensive score;

[0090] The formula balances the interaction effect of environment score and user behavior intensity by considering the weight of environment score and user behavior intensity on user's actual behavior respectively. It represents the interaction integral of the environment score and the user behavior intensity over time, thereby improving the scoring effect of the comprehensive score S.

[0091] By combining environmental scores, user behavior intensity and the interaction between the two, a more comprehensive integrated score is provided, which can more accurately predict user travel behavior and take into account users' personalized behavior patterns and environmental impacts, so that the model can maintain stable prediction performance in different environments and behavioral characteristics, and is more adaptable. At the same time, the use of comprehensive scores can better utilize public service data resources, and based on the impact on the score, assist in, for example, traffic scheduling and public service arrangements, improve the efficiency of the overall system and user experience, and combine personal behavior data with public service data to provide a scientific basis for urban planning and policy making, and achieve the effect of comprehensive analysis of public service data and identification of user behavior based on artificial intelligence, providing users with travel predictions and recommendations, providing users with psychological expectations for travel, reducing the impact on user behavior, and reducing the pressure on traffic and environmental public services.

[0092] Furthermore, the threshold is dynamically optimized to determine travel recommendations based on the comprehensive score, including:

[0093] Calculate the statistics of the comprehensive score S in the user's historical travel data, including the mean, standard deviation and median, and add the mean and standard deviation as the preliminary threshold ;

[0094] Based on historical travel data, the travel ratio TPR of actual travel users that are correctly predicted is calculated, and the non-travel ratio TNR of non-travel users that are correctly predicted is calculated;

[0095] Adjust the threshold gradually And verify that the correct prediction ratio of TNR and TPR reaches the maximum, determine the threshold ;

[0096] According to the determined threshold If the comprehensive score S is greater than the threshold, the user is recommended to travel. If the comprehensive score S is less than or equal to the threshold, the user is not recommended to travel.

[0097] By combining user historical behavior and feedback, data support is provided for threshold optimization, so that threshold-based judgments can provide users with more personalized travel suggestions. By optimizing the threshold, the model can more accurately predict the user's travel decisions, achieve scientific determination of the threshold, and provide a reliable basis for model optimization. Through accurate travel predictions, it helps public service data and improves system efficiency and user experience.

[0098] The data management module performs hierarchical storage and backup based on the cloud service storage platform, automatically generates reports based on stored data, and regularly updates data;

[0099] Preferably, hierarchical storage and backup are performed based on a cloud service storage platform, including:

[0100] The historical data, user behavior data and comprehensive score data are remotely transmitted to the cloud service storage platform via wireless network, and the data is regularly backed up through the AWS S3 formula;

[0101] Use cloud storage services to store data separately and adopt a tiered storage strategy based on data access frequency and importance;

[0102] AWS S3 is used as cold data storage for historical data and backup data, and AWS S3 is used as hot data storage for real-time comprehensive scores and user behavior data.

[0103] Through the tiered storage and backup mechanism of cloud storage, efficient data storage and management are provided to ensure high availability and scalability of the system. Tiered storage can differentiate data storage access frequency and latency requirements, so that different data types can be used more efficiently while reducing the burden on cloud servers. By backing up data, the security of data storage can be improved, thereby improving the security of the use of the cloud service platform.

[0104] Furthermore, reports are automatically generated based on the stored data and the data is updated regularly, including:

[0105] Use AWS QuickSight based on the cloud service platform as a report generation tool, integrate the data in the cloud storage with the report generation tool, and build data models and reports;

[0106] Use AWS QuickSight to configure the automatic refresh policy of the SPICE dataset to synchronize data regularly and configure the automatic update function of the report generation tool to refresh data and reports regularly.

[0107] Through the reporting tools of cloud services, reports can be generated based on data to visualize the data, making it more intuitive and easy to understand, helping technical personnel to quickly understand the data content. At the same time, reports are automatically generated and updated, which improves data visualization and decision support capabilities, improves the timeliness of reports, and makes it convenient for technical personnel to view and understand at any time.

[0108] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0109] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0110] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit with a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit with a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0111] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A public service data analysis system based on artificial intelligence, characterized by: include, Data processing module, based on public service data and personal user information, and extracts features; Data analysis module, builds regression models, analyzes public service data to predict environmental scores, analyzes user travel behavior based on personal user information, and analyzes user behavior intensity; Comprehensive scoring module, which integrates environmental scores and user behavior intensity, analyzes the comprehensive score of travel, dynamically optimizes thresholds, and makes travel recommendations based on the comprehensive score; The data management module performs hierarchical storage and backup based on the cloud service storage platform, automatically generates reports based on stored data, and regularly updates data; The method is based on public service data and personal user information, and extracts features, including: Based on the feature extraction of environmental data and traffic data in public service data, including PM2.5, temperature, humidity, ultraviolet intensity and traffic flow data, the PM2.5 is logarithmically changed, and the temperature, humidity, ultraviolet intensity and traffic flow data are standardized to form the environmental feature vector E; Collect personal user information, including user location information at different times, travel displacement data, and historical travel time data, perform standardization processing, and form a user feature vector R; The analyzing of user travel behavior and the intensity of user behavior based on personal user information includes: Based on the user feature vector R, the user behavior intensity is calculated and expressed as: ； in represents the behavior intensity of the i-th user at time t, represents the adjustment coefficient of the intensity of the periodic behavior of the i-th user, T represents the time period, L represents the reference distance of the permanent location, Indicates the distance between the current location of user i and the permanent location, represents the time mean of user i’s current behavior, represents the time standard deviation of user i’s current behavior, represents the adjustment coefficient of user i’s historical action data, represents the time mean of the k-th historical behavior, represents the time standard deviation of the k-th historical behavior, is the length of the time period, K represents the total number of historical behavior events; By collecting historical data and calibrating the historical user behavior intensity data based on user behavior as training set data, setting the mean square error loss function, using the gradient descent method to minimize the loss function, adjusting and confirming and ; Use the newly collected user feature data to predict the user's behavior intensity; The said construction of regression model and analysis of public service data to predict environmental scores includes: Based on the environmental feature vector E, a regression model is constructed to perform environmental scoring for the target public service scenario, which is expressed as: ； in represents the predicted environmental score, Indicates the upper limit of the frequency division time interval, represents the weight coefficient of the i-th environmental feature, represents the adjustment coefficient of the i-th environmental feature, represents the frequency coefficient of the i-th feature, represents the i-th eigenvalue at time t, and N represents the number of features; Use the training set to train the model and set the mean square error loss function; Use gradient descent to minimize the loss function to update and determine the parameters , and ; Use the newly collected feature data to input the model to obtain the current environmental score.

2. The public service data analysis system based on artificial intelligence according to claim 1, characterized in that: The comprehensive environment score and user behavior intensity are used to analyze the comprehensive score of the trip, including: According to the predicted environment score and user behavior intensity, the comprehensive score is calculated and expressed as: ； Where S represents the comprehensive score, represents the environmental score adjustment coefficient, represents the user behavior intensity adjustment coefficient, The interaction coefficient between the environment score and the user behavior intensity; By collecting historical data, we define historical comprehensive scores as training sets based on different weather environment data and users' actual behaviors, use mean square error as the objective function and define the parameter range, and divide the data set into k subsets through k-fold cross validation. Each time, we use k-1 subsets for training and the remaining subset for validation. This is repeated k times, and the average value is taken as the value of the objective function. The value that minimizes the objective function is used to confirm the result. , and ; Using the newly predicted environment score and the intensity of the user's behavior, a comprehensive score is calculated.

3. The public service data analysis system based on artificial intelligence as claimed in claim 2, characterized in that: The dynamic optimization threshold determines the travel recommendation based on the comprehensive score. include, Calculate the statistics of the comprehensive score S in the user's historical travel data, including the mean, standard deviation and median, and add the mean and standard deviation as the preliminary threshold ; Based on historical travel data, the travel ratio TPR of actual travel users that are correctly predicted is calculated, and the non-travel ratio TNR of non-travel users that are correctly predicted is calculated; Adjust the threshold gradually And verify that the correct prediction ratio of TNR and TPR reaches the maximum, determine the threshold ; According to the determined threshold If the comprehensive score S is greater than the threshold, the user is recommended to travel. If the comprehensive score S is less than or equal to the threshold, the user is not recommended to travel.

4. The public service data analysis system based on artificial intelligence as claimed in claim 3, characterized in that: The layered storage and backup based on the cloud service storage platform includes: The historical data, user behavior data and comprehensive score data are remotely transmitted to the cloud service storage platform via wireless network, and the data is regularly backed up through the AWS S3 formula; Use cloud storage services to store data separately and adopt a tiered storage strategy based on data access frequency and importance; AWS S3 is used as cold data storage for historical data and backup data, and AWS S3 is used as hot data storage for real-time comprehensive scores and user behavior data.

5. The public service data analysis system based on artificial intelligence as claimed in claim 4, characterized in that: The automatic generation of reports based on stored data and regular updating of data include: Use AWS QuickSight based on the cloud service platform as a report generation tool, integrate the data in the cloud storage with the report generation tool, and build data models and reports; Use AWS QuickSight to configure the automatic refresh policy of the SPICE dataset to synchronize data regularly and configure the automatic update function of the report generation tool to refresh data and reports regularly.

Images